While radiotherapy has made some impact on increased median survival for patients with high-grade glic brain metastases, these patients still generally progress locally, indicating the need for improved means control. The use of chemotherapy in combination with radiation has not shown significant improvement it is possible that the limited added benefit of chemotherapy is in part due to the large molecular size of inhibiting transport across the blood-brain barrier (BBB). Findings from radiation toxicity studies, however, reveal that radiation may loosen endothelial tight junctions, thus increasing transport of large molecules to irradiate tissue. As the dose to the tumor region is markedly higher than that delivered to most normal brain, this effect fact increase specificity of uptake of administered agents. This may provide a window of opportunity to improve efficiency of delivery of chemotherapeutic agents or radiosensitizers to tumor cells during radiotherapy. The of this radiation effect needs to be better understood to determine potential advantage, as to date, the time dose and volume dependence, and time course of repair are not well understood. This study, in response to a for applications on In Vivo Cancer Imaging Exploratory/Developmental grants from NCI/NIH/DHHS (PA-04-045 proposes to systematically characterize these dependencies on vascular permeability of normal brain and tumor to a MRI contrast agent that has a molecule size similar to many chemotherapeutic agents. Patient's und radiotherapy for high-grade gliomas or brain metastases will be followed by a sequence of dynamic c enhanced (DCE) MRI scans. Data from these scans will be analyzed to quantify vascular permeability, which related to dose, volume, time, and presence in tumor versus normal brain tissue.